How is the proper pattern of neural connections established during development? And how is that pattern maintained in the adult nervous system? These are the questions that the Axon Guidance and Neural Connectivity Unit seeks to answer. There is a great deal of interest in developing models of neurodegenerative diseases in simple, invertebrate model systems that provide unequaled experimental power for characterizing the cellular events of a complex process and establishing its molecular genetic basis. Use of Drosophila for studies of neurodegeneration have been problematic, however, since in general they have either relied on highly artificial manipulations, such as high-level expression of mutated human genes in the fly, or have identified genes that clearly affect neuronal survival in the fly but are not related to any gene or pathway demonstrated to play a role in neurodegeneration in mammals. We have now identified a natural, adult-onset neurodegenerative syndrome of Drosophila in flies mutant for the ortholog of a gene directly implicated in human diseases including Alzheimer Disease and ALS. The protein kinase Cdk5, together with its regulatory subunit, p35, is one of the major kinases that phosphorylates cytoskeletal proteins to generate the neurofibrillary tangles that are characteristic of the "tauopathy" class of neurodegenerative diseases. Moreover, activated Cdk5 is found concentrated in degenerating tissue in the brains of Alzheimer patients, and experimental activation of Cdk5 induces degenerating lesions in the mouse brain. We have generated a null mutation of the gene encoding the fly homolog of the Cdk5 activating subunit, p35. We find the mutants are viable and fertile, and are behaviorally normal at birth. However, within a few weeks they show progressive loss of motor coordination, culminating in rigidity and then death, with a significantly shortened lifespan (30% shorter than matched controls). Sectioning the heads of aging p35 mutant flies reveals degenerative lesions in the brain, initially in the neuropil but also around the cell bodies. Remarkably, these lesions are highly localized, being present bilaterally in specific brain nuclei, but not generally distributed through the brain, even though p35 and Cdk5 are present and active throughout the brain. Therefore, these mutants may allow us not only to uncover the genetic pathway leading to Cdk5-associated neurodegeneration, but also to understand how and why disease processes that occur throughout the brain lead to very specific and characteristic structural and behavioral defects in particular brain regions. Moreover, in mutant animals we observe widespread defects in axon patterning, synaptic morphology and protein localization within axons long before we see overt degeneration, raising the possibility that late onset degeneration may actually reflect a delayed response to defective nervous system structure early in development.